Two corpus sources agree that gene-fitness data contradict a curated functional annotation of ManX, a component of the mannose phosphotransferase system (PTS, a sugar uptake system that phosphorylates sugars as it imports them). 1,2 They do not agree on the per-substrate fitness values behind that conclusion. 1,2 This matters for evidence-triangulation-for-functional-annotation: if fitness evidence is to overrule a curated annotation, the numbers it rests on must be traceable. Here, the central digest and the project report give different values for the same genes and conditions.
Evidence Sides
Discoveries digest
The digest draws on Fitness Browser data from 168 experiments with the E. coli K-12 Keio collection, a library of single-gene knockouts. 1 It reports severe ManXYZ knockout defects on D-mannose (fit = -3.93) and D-glucosamine (fit = -2.75). 1 It reports dispensability on D-fructose (fit = +0.18 to +0.66). 1 On that basis it says the data contradict the UniProt annotation “fructose import across plasma membrane” (GO:0005354, a Gene Ontology term) for ManX. 1
snipe_defense_system project report
The report gives per-gene values. On D-mannose they are manX -2.75, manY -3.00 and manZ -2.74. 2 On D-glucosamine they are manX -3.79, manY -2.79 and manZ -3.63. 2 On D-fructose they are manX -0.22, manY +0.15 and manZ -0.07. 2 The report identifies -3.93 as manX’s worst fitness across 168 experiments, not as its D-mannose value. 2 It presents its table as Fitness Browser data that directly contradict UniProt’s annotation of ManX as a “fructose-specific” PTS component. The table shows ManXYZ defects on D-mannose and D-glucosamine but not on D-fructose, where FruA fitness is -1.44. 2
Shared ground
Both sources report defects on mannose and glucosamine and no defect on fructose. The qualitative challenge to the GO annotation is therefore common to both. 1,2 The digest’s substrate-level values do not match the project report, and they should not be cited until the two are reconciled. 1,2
Possible Reconciliations
- Hypothesis: the digest’s D-mannose figure (-3.93) 1 may be manX’s worst fitness across 168 experiments, which the report gives as a summary value rather than a D-mannose value 2, mislabelled as a substrate value.
- Hypothesis: the digest’s D-glucosamine figure (-2.75) 1 may have been transposed from the report’s manX D-mannose value, which is also -2.75. 2
- Hypothesis: the digest’s D-fructose range (+0.18 to +0.66) 1 may come from a different gene, replicate set or aggregation than the report’s per-gene values of manX -0.22, manY +0.15 and manZ -0.07. 2 Neither source states which.
None of these is confirmed by the evidence. Each would need checking against the underlying records.
Resolving Work
- Re-query the Fitness Browser E. coli K-12 records for manX, manY and manZ on D-mannose, D-glucosamine and D-fructose. Tabulate per-experiment and averaged fit values to establish which source’s substrate-level numbers the data reproduce.
- Trace the digest’s provenance by comparing its figures against the snipe_defense_system tables, testing whether a worst-fitness value or a transposed cell was copied into the digest.
- Recover the replicate-level D-fructose scores for all three genes, showing whether any experiment or aggregation yields the digest’s +0.18 to +0.66 range. 1
- Once the values are settled, correct or annotate the digest entry and re-state the GO:0005354 contradiction with cited per-gene values, so that triangulation pages cite one consistent table.